Neural Arabic Text Diacritization: State-of-the-Art Results and a Novel Approach for Arabic NLP Downstream Tasks

In this work, we present several deep learning models for the automatic diacritization of Arabic text. Our models are built using two main approaches, viz. Feed-Forward Neural Network (FFNN) and Recurrent Neural Network (RNN), with several enhancements such as 100-hot encoding, embeddings, Conditional Random Field (CRF), and Block-Normalized Gradient (BNG). The models are tested on the only freely available benchmark dataset and the results show that our models are either better or on par with other models even those requiring human-crafted language-dependent post-processing steps, unlike ours. Moreover, we show how diacritics in Arabic can be used to enhance the models of downstream NLP tasks such as Machine Translation (MT) and Sentiment Analysis (SA) by proposing novel Translation over Diacritization (ToD) and Sentiment over Diacritization (SoD) approaches.

AN APPROACH TO EVALUATION OF FREIGHT TRAINS BRAKING EFFICIENCY BY THE RESULTS OF EXPERIMENTAL STUDIES IN COMPLIANCE WITH HOST 34434-2018 REQUIREMENTS

The approach of experimental study of the freight trains braking efficiency using computer simulation based on the implementation of the simulation model in the form of a differential equation of the wagon motion during braking, is proposed. The methodology of experimental studies is based on universal formulas for power-law dependences of braking parameters. The braking efficiency is evaluated by the use of computer software packages written in VBA (Visual Basic for Application) in Excel. The software package al-lows you to increase the automation of testing, the accuracy of calculations, to reduce the time for testing, and also minimize the number of errors caused by the human factor. The presented methodology significantly extends the number of parameters of the braking process used to analyze the braking efficiency of a freight train based on the results of running braking tests: actual values of braking coefficients; braking distances of a freight train not only on the site, but also on normalized gradient descents for a given number of wagons in the train, taking into account the increase in the braking force along the wagons in the train formation; actual values of wheel and rail adhesion coefficients during braking; the deceleration of the freight wagon and the train during braking, as well as the braking time. Examples of computational and experimental studies of a freight train with tread brake are given. A comparative analysis of experimental and computational studies demonstrates sufficiently satisfactory matching of their results. Key words: freight train, braking distance, braking speed, braking coefficient, increase in braking force, mathematical model, trend line, braking wave.